Role of histamine H3 receptor in glucagon-secreting αTC1.6 cells.

Pancreatic α-cells secrete glucagon to maintain energy homeostasis. Although histamine has an important role in energy homeostasis, the expression and function of histamine receptors in pancreatic α-cells remains unknown. We found that the histamine H3 receptor (H3R) was expressed in mouse pancreatic α-cells and αTC1.6 cells, a mouse pancreatic α-cell line. H3R inhibited glucagon secretion from αTC1.6 cells by inhibiting an increase in intracellular Ca(2+) concentration. We also found that immepip, a selective H3R agonist, decreased serum glucagon concentration in rats. These results suggest that H3R modulates glucagon secretion from pancreatic α-cells.

Introduction

Histamine has multiple functions; it is released during allergic reactions, affects gastric acid secretion, and impacts the sleep–wake cycle. These functions are mediated thorough interactions with four different subtypes (H1–H4), which belong to the G protein-coupled receptor superfamily. Among these subtypes, histamine H3 receptor (H3R) is currently a hot topic because of its important role in energy homeostasis [1]. Neuronal histamine has an inhibitory effect on food intake through modulation of hypothalamic neuronal activities [2,3]. H3R expressed in the central nervous system (CNS) regulates histamine release [4]. Therefore, H3R is a prospective therapeutic target for obesity, which could potentially lead to reduced body weight [5].

The regulation of glucagon secretion from pancreatic α-cells is fundamental to the maintenance of glucose homeostasis; under normal conditions they secrete glucagon in response to starvation-induced hypoglycemia. Glucagon is a 29-amino acid peptide hormone that elevates blood glucose concentration by glycogenolysis and gluconeogenesis in the liver [6]. Glucagon secretion from pancreatic α-cells has recently become a hot topic as a new therapeutic target for diabetes mellitus [7], as dysfunctional pancreatic α-cells have been reported in diabetic patients [8]. In these patients, blood glucagon concentrations are elevated even in hyperglycemic states [9]. Conversely, glucagon secretion is insufficient in response to critical hypoglycemia [10]. This inappropriate glucagon secretion aggravates diabetes. Therefore, it is crucial to understand the mechanism of glucagon secretion.

The key secretagogue for glucagon is low serum glucose concentration [11]. Low glucose concentrations facilitate glucagon release from pancreatic α-cells through the activation of voltage-dependent Ca2+ channels (VDCCs) [12]. In addition to glucose, many paracrine/endocrine products and neurotransmitters are involved in the modification of glucagon secretion [13]. For example, glucagon-like peptide (GLP)-1, an incretin secreted from the small intestine L-cells, decreases glucagon secretion from pancreatic α-cells [14].

There are some biological molecules that modulate pancreatic α-cells and CNS functions to regulate total energy homeostasis. GLP-1 has an important role in regulating eating behavior by modulating CNS activities. Turton et al. reported that administration of GLP-1 into the lateral ventricle resulted in inhibition of eating behavior [15]. This evidence indicates that pancreatic α-cells and CNS are targets of GLP-1. Additionally, leptin, a peptide adipokine secreted from peripheral adipose tissues [16], has inhibitory effects on pancreatic α-cells [17] and food intake [18]. Moreover, ghrelin, a 28-amino acid peptide synthesized in the gastrointestinal tract, directly stimulates glucagon secretion [19] and eating behavior [20]. As previously mentioned, hypothalamic H3R has an important role in food intake; thus, it is interesting to examine whether H3R is expressed in pancreatic α-cells and involved in the regulation of energy homeostasis.

Taking this evidence into account, we hypothesize that H3R is expressed and involved in pancreatic α-cell function. We used a mouse pancreatic α-cell line, αTC1.6 cells, to evaluate the function of H3R in pancreatic α-cells. These αTC1.6 cells have properties similar to those of pancreatic α-cells, and they are widely used in research on glucagon secretion [21]. We discovered inhibitory effects of H3R on glucagon secretion in vitro and in vivo using pharmacological assays.

Methods

Animals

Wild-type (WT) and histamine H3 receptor-gene KO (H3KO) C57BL/6 [22] mice were bred in our laboratory. Wister rats were purchased from Japan SLC (Hamamatsu, Japan). All mice were maintained on a 12-h light/dark cycle in a humidity- and temperature-controlled room, and they were allowed ad libitum access to food and water. This study was approved by the Center for Laboratory Animal Research, Tohoku University, Sendai, Japan. This study had been performed in accordance with the EU Directive 2010/63/EU for animal experiments.

Cell culture

αTC1.6 cells were purchased from the American Type Culture Collection (Manassas, VA, USA). Cells were maintained in DMEM (Invitrogen, Carlsbad, CA, USA) containing 10% FCS (Thermo, Waltham, MA, USA), 15 mM HEPES (Gibco, Carlsbad, CA, USA), 0.1 mM non-essential amino acids (Gibco), 0.02% BSA (Sigma, St. Louis, MO, USA), 1.5 g/dL sodium bicarbonate (Sigma), 2 g/L d-glucose (Sigma), 75 mg/mL penicillin G potassium (Wako, Osaka, Japan), and 100 mg/L streptomycin sulfate (Wako) at 37 °C in a 5%-CO2 humidified incubator.

RT-PCR

RT-PCR was performed using a previously described protocol [23]. Briefly, total RNA was isolated from αTC1.6 cells and reverse-transcribed. Samples (equivalent to 5 ng total RNA) were amplified by 35 cycles of PCR (10 s at 98 °C, 30 s at 60 °C, and 90 s at 72 °C).

Western blot

Western blot of H3R was performed using a previously described protocol [23]. We used the hypothalamic lysate from WT and H3KO mice as positive and negative controls, respectively. Beta-actin was detected by an antibody (Cell Signaling Technology, Danvers, MA, USA) as a loading control.

Immunohistochemical analysis

Pancreata were quickly removed from 12-week-old WT and H3KO mice. Isolated pancreata were fixed in 10% paraformaldehyde and embedded in paraffin. The paraffin sections were deparaffinized and blocked with 10% goat serum at room temperature for 15 min. To detect glucagon, sections were incubated with mouse anti-glucagon antibody (Sigma; 1:2000 dilution) for 2 h at room temperature and then treated with the secondary antibody: Alexa Fluor (Invitrogen) 568 conjugated goat anti-mouse antibody (1:100 dilution). To detect H3R, samples were incubated overnight at 4 °C with rabbit anti-mouse H3R antibody (4 μg/mL) [23] and then treated with the secondary antibody: Alexa Fluor 488 conjugated goat anti-rabbit antibody (Invitrogen). For double staining of glucagon and H3R, samples were incubated with anti-glucagon antibody for 2 h at room temperature, washed thrice with PBS, followed by overnight incubation with anti-H3R antibody at 4 °C. Specimens from H3KO mice were used as negative controls for H3R. Finally, all samples were encapsulated with mounting medium containing DAPI.

Glucagon secretion assay

αTC1.6 cells were seeded into a 24-well plate at a density of 2.0 × 105 cells per well and then preincubated in 500 μL of Krebs–Ringer bicarbonate buffer (KRB) containing 20 mM glucose for 30 min at 37 °C in 5% CO2. Subsequently, αTC1.6 cells were incubated in 500 μL of KRB containing 20 or 2.8 mM glucose ± 1 μM histamine (Sigma), 1 and 100 μM of the selective H3R agonist immepip (Sigma) [24,25], or the selective H3R inverse agonist JNJ-5207852 (kindly gifted from Dr. Nicholas Carruthers, Johnson & Johnson Pharmaceutical Research and Development, USA) [26,27], or KRB containing 20 mM KCl ± 1 μM immepip, for 1 h at 37 °C in 5% CO2. Immepip and JNJ-5207852 were dissolved in distilled water. We used a glucagon ELISA kit (Wako) to measure glucagon concentration in KRB.

Intracellular Ca2+concentration measurement

αTC1.6 cells were seeded into a 96-well plate (Greiner, Frickenhausen, Germany) at a density of 2.0 × 104 cells per well and were then preincubated in KRB containing 20 mM glucose, 5 μM fluo-4/AM, 5 mM probenecid, and 1 × Power load (all from Invitrogen) for 30 min at 37 °C. Preincubated KRB was replaced with KRB containing 20 mM glucose ± 1 μM immepip and the 96-well plate was placed into a microplate reader FlexStation 3 (Molecular Devices, Sunnyvale, CA, USA) to record changes in fluorescence (excitation: 495 nm/emission: 518 nm). Cells were excited at 495 nm and the light emitted through a 518-nm filter was detected. After a 1-min measurement of baseline fluorescence, glucose concentration in KRB in each well was diluted to 2.8 mM by adding KRB ± 1 μM immepip from injectors, or KRB containing high KCl ± 1 μM immepip was added from injectors to adjust KCl concentration to 20 mM. Fluorescent intensity before fluo-4/AM loading was measured as background.

Measurement of serum glucagon concentration in rats

We used rats instead of mice to evaluate the effect of immepip on serum glucagon concentration because a glucagon ELISA kit required 100 μL of serum to measure glucagon correctly. 10-week-old Wister rats were fasted for 24 h prior to the experiment and were anesthetized with isoflurane. Immepip (30 mg/kg) dissolved in normal saline or just normal saline was administered intraperitoneally. Blood samples were collected from the caudal vein before injection as the baseline and at 15 min after injection. Samples were centrifuged for 30 min at 5000 rpm at 4 °C and then the supernatant plasma was collected. Plasma glucagon concentration was measured using a glucagon ELISA kit (Wako).

Data analysis and presentation

All experiments were performed in triplicate. Statistical analysis was performed using Student’s t-test or one-way ANOVA. Data are expressed as mean ± SE. P-values of <0.05 were considered significant.

Results

H3R were expressed in αTC1.6 cells

First, we examined the expression of histamine receptors in αTC1.6 cells using RT-PCR and western blot. H3R mRNA was predominantly expressed in αTC1.6 cells among the four histamine receptor subtypes (Fig. 1A). Histamine H1 receptor mRNA was slightly detected, but histamine H2 and H4 receptor mRNAs were not detectable. H3R protein expression in αTC1.6 cells was confirmed by western blotting (Fig. 1B). Immunohistochemical analysis of pancreatic islets showed that glucagon-positive (stained red) expressed H3R protein (stained green), indicating that H3R is expressed in glucagon-positive mouse pancreatic α-cells (Fig. 1C). H1R protein expression was not detectable during western blot of αTC1.6 cells (data not shown).

Fig. 1

Histamine H3 receptor expression in αTC1.6 cells and pancreatic islets. (A) RT-PCR detection of histamine receptors in αTC1.6 cells. RT + and − indicate RT-PCR in the presence or absence of reverse transcriptase, respectively. H1R: histamine H1 receptor, H2R: histamine H2 receptor, H3R: histamine H3 receptor, H4R: histamine H4 receptor. (B) Western blot of H3R in αTC1.6 cells. “WT” is a wild-type mouse hypothalamus lysate used as a positive control. H3KO is a negative control, a hypothalamus lysate from an H3R-knockout mouse. Beta-actin (Actb) was used as a loading control (lower panel). (C) Immunohistochemical analysis of H3R expression in mouse pancreatic islets of WT mice (upper panels) and H3KO mice (lower panels) as negative controls. Glucagon and H3R were visualized with Alexa 568 (red) and Alexa 488 (green), respectively. DAPI is depicted in merged images (right panels). Scale bar = 50 μM. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Activation of H3R decreased glucagon secretion

Glucagon secretion from pancreatic α-cells is vital to enhance serum glucose concentration in hypoglycemia [6]. We investigated effects of histamine on glucagon secretion from αTC1.6 cells. Glucagon secretion increased 1.41-fold over basal concentration in response to low glucose concentrations (Fig. 2A). This secretion associated with low glucose concentrations was remarkably inhibited by 1 μM histamine (0.67 ± 0.05 ng/well/h versus 0.96 ± 0.07 ng/well/h, P < 0.05), while basal secretion at 20 mM glucose with histamine was maintained at close to control concentrations (0.68 ± 0.06 ng/well/h versus 0.68 ± 0.06 ng/well/h, P > 0.05; Fig. 2A). These results indicate that histamine negatively regulates glucagon secretion from αTC1.6 cells in response to low glucose concentration.

Fig. 2

Effects of histamine, immepip, or JNJ-5207852 on glucagon secretion from αTC1.6 cells. (A) Glucagon secretion from αTC1.6 cells at low (2.8 mM) and high (20 mM) glucose concentrations ± 1 μM histamine. ∗P < 0.05, n = 6. (B) Glucagon secretion from αTC1.6 cells at low (2.8 mM) and high (20 mM) glucose concentrations ± 1 or 100 μM immepip. ∗P < 0.05, n = 6. (C) Glucagon secretion from αTC1.6 cells at low (2.8 mM) and high (20 mM) glucose concentrations ± 1 or 100 μM JNJ-5207852. ∗P < 0.05, n = 6.

Subsequently, we examined the role of H3R expressed in αTC1.6 cells in glucagon secretion using pharmacological assays with the H3R agonist immepip or the inverse agonist JNJ-5207852. Glucagon secretion associated with low glucose concentrations was remarkably inhibited by 1 μM immepip (0.73 ± 0.08 ng/well/h versus 1.01 ± 0.10 ng/well/h, P < 0.05; Fig. 2B). Conversely, 1 μM JNJ-5207852 increased glucagon secretion (1.83 ± 0.08 ng/well/h versus 1.36 ± 0.21 ng/well/h, P < 0.05; Fig. 2C). These results indicate that H3R inhibits glucagon secretion in αTC1.6 cells.

Immepip inhibited the increase of intracellular Ca2+ concentration

H3R activation modulates cellular functions through several intracellular signaling pathways downstream of H3R [28,29]. Silver et al. reported that H3R activation attenuated norepinephrine exocytosis mediated by a decrease in intracellular Ca2+ concentration, seemingly because of an impaired entrance of Ca2+ through VDCCs [30]. Glucagon exocytosis from pancreatic α-cells also requires an increase in intracellular Ca2+ concentration [31,32]. Thus, we examined the effect of activated H3R on intracellular Ca2+ concentration. One micromole of immepip impaired the increase in intracellular Ca2+ concentration in response to 2.8 mM glucose (transient changes in Ca2+; 0.36 ± 0.03 versus 0.47 ± 0.04; Fig. 3A). This result indicates that H3R is involved in the regulation of intracellular Ca2+ concentration in αTC1.6 cells.

Fig. 3

Analysis of intracellular signaling in αTC1.6 cells. (A) Changes in intracellular Ca2+ concentration in response to low (2.8 mM) glucose concentration ± 1 μM immepip. The average of fluorescence intensity from fluo-4 before low glucose stimulation was arbitrarily set as 1. Intensity ratio on y-axis relates to the increase of intracellular Ca2+ concentration in direct proportion. 20 G: 20 mM glucose, 2.8 G: 2.8 mM glucose, n = 4. (B) Glucagon secretion from αTC1.6 cells at high (20 mM) glucose concentration and high (20 mM) KCl concentration ± 1 μM immepip. ∗P < 0.05, n = 6. (C) Changes in intracellular Ca2+ concentration in response to high (20 mM) KCl concentration ± 1 μM immepip. The average of fluorescence intensity from fluo-4 before low glucose stimulation was arbitrarily set as 1. Intensity ratio on y-axis relates to the increase of intracellular Ca2+ concentration in direct proportion. 20 G: 20 mM glucose, 20 K: 20 mM KCl, n = 4.

To confirm whether intracellular Ca2+ concentration has a direct role in the inhibitory effect of H3R on glucagon secretion, we stimulated αTC1.6 cells with 20 mM KCl solution, which directly depolarized plasma membranes, followed by Ca2+ influx through activated VDCCs [33]. We examined the effect of immepip on glucagon secretion stimulated by 20 mM KCl and found that glucagon secretion in response to 20 mM KCl was reduced by 1 μM immepip (1.82 ± 0.07 ng/well/h versus 2.29 ± 0.15 ng/well/h; Fig. 3B). This result suggests that intracellular signaling from H3R affects the increase in intracellular Ca2+ concentration and/or downstream Ca2+ signaling, so we examined effects of immepip on the increase in intracellular Ca2+ concentration in response to 20 mM KCl. One micromole of immepip impaired the increase in intracellular Ca2+ concentration in response to 20 mM KCl (transient changes in Ca2+; 0.13 ± 0.07 versus 0.55 ± 0.08; Fig. 3C). These results indicate that H3R signaling exerted an inhibitory effect on glucagon secretion through the blockage of Ca2+ influx.

H3R agonist reduced serum glucagon concentration in rats

Finally, we investigated the role of H3R in glucagon secretion in vivo. We examined the effect of immepip on serum glucagon concentration in rats. Serum glucagon concentration after a 24-h fast was remarkably inhibited by 1 μM immepip (75.3 ± 14.5 pg/mL versus 130.0 ± 12.8 pg/mL, P < 0.05; Fig. 4), indicating that H3R was involved in the regulation of serum glucagon concentration in vivo.

Fig. 4

Effects of immepip on rat plasma glucagon concentration. Serum glucagon concentration of rats before and 15 min after intraperitoneal injection of 30 mg/kg immepip. ∗P < 0.05, n = 4.

Discussion

We demonstrated for the first time that H3R is expressed in αTC1.6 cells and mouse pancreatic α-cells. Using RT-PCR and western blot analysis, αTC1.6 cells were shown to express H3R mRNA and protein (Fig. 1A, B). H3R protein expression in pancreatic α-cells was confirmed by immunohistochemical analysis. Our data showed that activated H3R inhibited glucagon secretion from αTC1.6 cells in response to low glucose concentrations (Fig. 2).

H3R is coupled to Gi protein [34]. Previous reports revealed that several Gi protein-coupled receptors expressed in pancreatic α-cells inhibited glucagon secretion. For example, group III metabotropic glutamate receptors, coupled with inhibitory Gi protein, are expressed in rat pancreatic α-cells and inhibit glucagon secretion from rat islets under low glucose conditions [35]. Moreover, somatostatin from pancreatic δ-cells also inhibits glucagon secretion through somatostatin receptor 2 coupled to Gi proteins [36]. This evidence corroborates the function of H3R in pancreatic α-cells.

To elucidate the regulatory mechanism of H3R on glucagon secretion, we investigated effects of immepip on intracellular Ca2+ concentration, which is essential to glucagon secretion [13]. Immepip inhibited increases in intracellular Ca2+ concentrations in response to 20 mM KCl stimulation (Fig. 3B). KCl stimulation directly elicits plasma membrane depolarization, leading to the activation of VDCCs. It is possible that H3R signaling might inhibit the activity of VDCCs in pancreatic α-cells because H3R in neurons reduced Ca2+ influx through VDCCs [37]. However, the inhibitory effect of immepip on Ca2+ concentration was more moderate during low glucose conditions than during KCl stimulation (Fig. 3B, C), although immepip inhibited glucagon secretion during low glucose conditions more strongly than during KCl stimulation (Figs. 2B and 3A). This suggests that the intracellular Ca2+-independent signaling pathway might be associated with the inhibitory effect of H3R on glucagon secretion. Recently, De Marinis showed that the cAMP-PKA pathway increased glucagon secretion by the recruitment of glucagon granules toward the cell membrane [38]. Activated H3R inhibits the cAMP-PKA pathway [24,39]; thus, any involvement of this pathway might be confirmed by using gene silencing technologies or a pharmacological approach. Further studies are necessary to elucidate the signaling induced by H3R stimulation in pancreatic α-cells.

Immepip remarkably inhibited plasma glucagon concentration 15 min after intraperitoneal injection, indicating that H3R has an important role in glucagon concentration in vivo. Our data suggests that immepip acts on pancreatic α-cells leading to direct inhibition of glucagon secretion. Additionally, immepip might indirectly modulate glucagon secretion through sympathetic nerves. Sympathetic nerve endings innervate pancreatic α-cells and modulate glucagon secretion [32,40]. Schlicker et al. reported that postganglionic sympathetic nerves expressed H3R and that this H3R regulates neurotransmitter release [41]. This evidence indicates that immepip might act on H3R in sympathetic nerves and indirectly regulate glucagon secretion. However, further studies are needed to elucidate the role of H3R in pancreatic α-cells or sympathetic nerves in the regulation of serum glucagon secretion.

As previously mentioned, abnormal glucagon secretion from pancreatic α-cells has been observed in diabetic patients [8]. H3R agonists that decrease glucagon secretion could improve excessive glucagon secretion in hyperglycemic conditions; however, our data only indicates the inhibitory effect of H3R on glucagon secretion in hypoglycemic conditions. Therefore, further studies are needed to clarify therapeutic effects of H3R agonists in diabetic animal models, which have abnormal glucagon secretion during hyperglycemia [42].

We found a weak band of H1R mRNA in αTC1.6 cells (Fig. 1A). Based on our previous study that clearly indicated the absence of H1R mRNA in pancreatic islets [23], its expression might be undetectably low or absent in pancreatic islets and it may be specific to the cell-line. Additionally, we confirmed that glucagon secretion was not affected by H1R-specific agonists (data not shown). Accordingly, it is suggested that H1R is not involved in the regulation of glucagon secretion in pancreatic α-cells.

In this study, we found that H3R in pancreatic α-cells had an inhibitory effect on glucagon secretion possibly by decreasing intracellular Ca2+ concentration. We also found that H3R was involved in glucagon secretion in vivo. It will be interesting to investigate roles of H3R in glucagon secretion in a diabetic animal model.

Conflict of interest

The authors declare no conflicts of interest.